MCP App Store
by KeenEthics
Brandefense Atlas icon

Brandefense Atlas MCP Server for Claude and ChatGPT

By Brandefense

Overview

Brandefense Atlas brings Brandefense's free threat-intelligence tools into Claude. Ask how exposed a domain is and Atlas runs a domain risk scan across breached credentials, infected devices, dark web and dark chat mentions, phishing look-alike domains, exposed documents and data sales, and returns letter grades with an overall score. Search the Atlas CVE catalogue by vendor, product or severity, with CVSS, EPSS and known-exploited (KEV) flags, or read one CVE in full. Check public-facing software (NetScaler, GitLab, Atlassian Data Center, WordPress, PHP, Node.js) for specific high-profile vulnerabilities; the checks only read public pages or compare versions and never send an exploit. After a scan you can ask Brandefense to e-mail the full report to a work address on that domain. No Brandefense account is needed: connecting takes one human check. Usage is limited per connection per day.

How to connect

  1. Claude: Open the Brandefense Atlas listing in the Claude connectors directory and choose Connect. Claude listing Brandefense Atlas exposes 11 tools here.
  2. ChatGPT: Open the Brandefense Atlas listing in the ChatGPT app directory and choose Connect. ChatGPT listing Brandefense Atlas exposes 11 tools here.

Tools

atlassian_vulnerability_check

ChatGPT
Check an Atlassian Data Center instance for CVE-2026-21589 Passively read the version a visitor-supplied domain, IP or URL prints on its public sign-in page or application-link manifest (Confluence, Jira, Bitbucket, Bamboo, Crowd, Fisheye, Crucible) and report whether it falls in the affected range of CVE-2026-21589 (Atlassian CVSS 9.3). Ordinary GETs of public pages, never sends an exploit request and requests no file from the web root. Returns the verdict synchronously.

get_cve

ChatGPT
Read one CVE from the Atlas catalogue by id (for example CVE-2023-40044): its description, scores, references and related CVEs of the same product. Use it when the user names a CVE or picks one from search_cves.

get_domain_scan

ChatGPT
Read an Atlas domain risk scan by its id: its status and, once completed, the full result. Use it to follow up on a scan_domain call that returned before the scan finished. Scan links expire after a few days.

gitlab_vulnerability_check

ChatGPT
Check a GitLab instance for CVE-2026-110264 Passively read the GitLab release a visitor-supplied domain, IP or URL shows on its public sign-in page and report whether it falls in the affected range of CVE-2026-110264 (CVSS 10.0). Ordinary GETs of one public page, never sends an exploit request. Returns the verdict synchronously.

netscaler_vulnerability_check

ChatGPT
Check a NetScaler for CTX697096 exposure Passively fingerprint the firmware build of a visitor-supplied NetScaler ADC / Gateway (by domain, IP, or URL) and report whether it predates the CTX697096 fix (CVE-2026-88771 / CVE-2026-88772). Reads only public pages and resources, never sends an exploit request. Returns the verdict synchronously.

nodejs_vulnerability_check

ChatGPT
Check a version against Node2Shell / HTTP2Shell Compare the output of node -v, or an exact Docker image tag, with the Node.js security release of 21 Oct 2026 for Node2Shell (CVE-2026-111497) and return a firm verdict. The optional stack field (default nodejs) switches to the same flaw, called HTTP2Shell, in the Brotli library, Bun, Python, Ruby, Envoy, ngx_brotli or Debian. An optional Express version adds an exposure note and never changes the verdict. Pure version comparison: no request is made to any server.

php_vulnerability_check

ChatGPT
Check a site for PHP2Shell exposure Passively read the PHP version a visitor-supplied domain, IP or URL advertises in its response headers and report whether it falls in the PHP2Shell affected range (CVE-2026-109618 / CVE-2026-109619). A single GET of the public homepage, never sends an exploit request. Returns the verdict synchronously.

request_full_report

ChatGPT
Ask Brandefense to e-mail the full Atlas domain risk report for a domain, the same request as the report form on the Atlas website. The report goes only to a work e-mail address on the scanned domain; personal webmail is refused. Before calling, show the user this consent text and the privacy policy, and set consent=true only if they agree: "I agree to receive the report and related communications from Brandefense. You can unsubscribe at any time. See our Privacy Policy." Privacy Policy: https://brandefense.io/privacy-policy/. Requests are limited per connection per day, and a refused request still counts.

scan_domain

ChatGPT
Run an Atlas domain risk scan: how exposed an organisation's domain is across breached credentials, infected devices, dark web and dark chat mentions, phishing look-alikes, exposed documents and data sales, with letter grades and an overall score. Use it when the user asks how risky, exposed or breached a company or domain is. Waits up to about 90 seconds for the result; if the scan is still running, call get_domain_scan with the returned id later. A repeat scan of the same domain returns the recent result (reused=true) and costs nothing, unless that result came back degraded (some sources failed): then a fresh scan runs and is counted. Fresh scans are limited per rolling 24 hours, and fresh_scans_left_today says how many remain.

search_cves

ChatGPT
Search the Atlas CVE catalogue: the vulnerabilities Atlas tracks, with severity, CVSS and EPSS scores and whether they are known to be exploited (KEV). Use it when the user asks about vulnerabilities in a product or vendor, or wants the most urgent ones. q matches CVE id, vendor, product and title (every word must match); severity is one of info, low, medium, high, critical. Results are most urgent first.

wordpress_exposure

ChatGPT
Check a site for wp2shell exposure Passively fingerprint the WordPress version of a visitor-supplied domain or URL and report whether it falls in the wp2shell affected range (CVE-2026-63030 / CVE-2026-60137). Reads only the target's public readme and homepage, never sends an exploit request. Returns the verdict synchronously.

atlassian_vulnerability_check

Claude

get_cve

Claude

get_domain_scan

Claude

gitlab_vulnerability_check

Claude

netscaler_vulnerability_check

Claude

nodejs_vulnerability_check

Claude

php_vulnerability_check

Claude

request_full_report

Claude

scan_domain

Claude

search_cves

Claude

wordpress_exposure

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/brandefense-atlas"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/brandefense-atlas)